Choice is central to a wide variety of behavioral health problems, from those problems in which the choice leads to immediate harmful consequences, such as disruptive behavior in the developmentally disabled, or inattention by children with attention deficit disorder, to those problems in which the choice leads to delayed harmful consequences, such as, smoking, and drug and alcohol use, and obesity. Behavior modification therapy is a promising form of treatment for behavior disorders. Generally with this approach to treatment, therapists arrange increased reinforcement for appropriate behaviors, which are occurring at a low rate and, in so far as possible, decreased reinforcement for the problem behaviors, which are occurring at a high rate. The proposed research examines a new model of choice that has implications for interventions to minimize or cure a wide variety of behavior health problems. Choice is modeled in the laboratory by presenting two response alternatives. The standard conception of this choice procedure is that the choice is between responding at the two alternatives. According to the at-the-alternative model, however, there are two sets of choices: At each alternative the fundamental choice is between staying and responding at the present alternative and switching to the other alternative. The at-the-alternative model predicts that only reinforcers obtained for switching from Alternative 1 to Alternative 2 will influence responding at Alternative 1; the generalized matching law predicts that reinforcers for staying at Alternative 2 also will influence responding at Alternative 1. Identifying the degree to which reinforcement at one alternative influences responding at the other alternative has implications for how behavioral interventions to treat behavior disorders are designed. Each year there will be 6 or 7 experiments conducted using rats as subjects in the existing psychology animal lab. The first aim of this research is to determine whether reinforcing behavior at one alternative also influences responding at the other alternative. The second aim is to extend the study of choice to choice among three alternative; extending the study of choice in this way is important because complex human behavior typically consists of choices among more than two alternatives. This research will help identify how to model choice in this more complex situation. The third aim of this research is to evaluate the data using the generalized matching law and the at-the-alternative model. Many behavior disorders, such as, alcohol and drug abuse, attention-deficit disorder with or without hyperactivity, and developmental disabilities result, in part, from the choices an individual makes. A goal of treatment is to help individuals make choices that have consequences that benefit the individual, rather than cause further harm. The proposed research will help in understanding how to design interventions that help individuals make better choices. [unreadable] [unreadable] [unreadable]